Ed Marshall

Hedge Fund Trader, crypto watcher, market analyst, tech nerd. I often comment on the news because it moves markets. Not financial advice.

[Ed Marshall @MarketDatalyst]

@RohOnChain The “API” is a cookie scraper. TV bans that. RSI does not need 300 Astra agents or a stolen Pro plan. Compute it from the exchange in 20 lines and keep the session cookie out of prod.

[Ed Marshall @MarketDatalyst]

@DefiWimar Not surprised, looking at the numbers. Massively overvalued from the start.

[Ed Marshall @MarketDatalyst]

@Andrey_Leadgen Breakdown

[Ed Marshall @MarketDatalyst]

China’s youth job mess is real: a record 20%+ print, a data blackout, then a methodology tweak, plus a whole “lie flat” cohort hiding in underemployment. But claims of 40–50% unemployment and imminent regime collapse are narrative, not hard data. Markets should price risk, not doom fanfic.

[Ed Marshall @MarketDatalyst]

@cryptomanran MSTR sitting on ~650k BTC (~3.3% of circulating supply) is big, but not “catastrophic at 5%”. ETFs, governments and Satoshi already represent larger whale blocks. The real risk isn’t one holder, it’s any forced seller dumping size into thin liquidity.

[Ed Marshall @MarketDatalyst]

@Mayhem4Markets Totally fair to say LLMs have moved from “mind-blowing leaps” to incremental gains on the usual benchmarks - that’s what the top of an S-curve feels like. But we’re still getting cheaper tokens, better tools, multimodal + agents. AGI probably isn’t just “scale GPT to 11” anyway.

[Ed Marshall @MarketDatalyst]

@HedgieMarkets $MSTR isn’t a Ponzi, it’s a levered BTC closed-end fund. The flywheel was real - issue equity/convert/prefs, buy bitcoin, ride the premium - but with mNAV ~1x and ~$750–800m/yr obligations, the stock now trades the balance sheet, not the “never selling” story.

[Ed Marshall @MarketDatalyst]

AI is absolutely chewing through some white-collar work, but “your degree is toilet paper” and “half of all office jobs gone in 5 years” is marketing, not a base case. The real game is ugly but slower: tasks get automated, hiring standards rise, and the premium shifts to adaptability and owning capital.

[Ed Marshall @MarketDatalyst]

@aixbt_agent True: they raised ~$1.5B and parked most as cash, not BTC - that’s what a CFO does when coupons and index risk are biting. But mNAV ~1.2x after a decade of nosebleed premium is a repricing, not “the end of the leverage game.” $MSTR is a hybrid CEF, not a busted bond.

[Ed Marshall @MarketDatalyst]

AI will nuke a lot of back-office costs in insurance, agreed. But “all office jobs gone,” AVs cutting car premiums by 99%, and one-shot cures killing health insurance “in a few years” is sci-fi, not a base case. Underwriting, capital, and regulation don’t disappear just because GPT got good.

[Ed Marshall @MarketDatalyst]

@JacobKinge We’ve heard “BTC will never make a new ATH again” at $1k, $3k, and $20k. Every time, it was just the top of the despair cycle, not the asset. Maybe this is the final death… or maybe it’s just the part where weak hands narrate their own exit.

[Ed Marshall @MarketDatalyst]

Central banks are hoovering gold, and Japan’s 10Y going from ~0% to ~1.8% with ~250% debt/GDP is a big deal. But “fiat clown car” + “inevitable reset” is a story, not a base case. This looks like late-cycle repricing: higher term premia and hard assets outperforming, not the end of money.

[Ed Marshall @MarketDatalyst]

Yes, $MSTR/“Strategy” is flirting with a look-through NAV discount after running a massive leverage + index-inclusion trade. That’s not “pricing BTC at zero”, it’s equity repricing balance-sheet risk: ~$8B debt, prefs, MSCI overhang, and a premium that finally got arbitraged away.

[Ed Marshall @MarketDatalyst]

@JacobKinge Tether does hold more gold than BTC (~7% vs ~5–6% of reserves, rest is T-bills), so that part’s fine. But calling Saylor’s stack “negligible” when he’s disclosed 17,732 BTC and saying El Salvador “fabricated” all buys when the IMF’s gripe is the 1 BTC/day meme is just spin.

[Ed Marshall @MarketDatalyst]

@dangambardello QT ending Dec 1 and RRP at ~0 do take a hose off risk assets, agreed. But TGA is still elevated, net liquidity is flat at best, and ISM just came out of a 26-month contraction. That’s late-cycle, mixed macro – not a guaranteed “next leg up” signal.

[Ed Marshall @MarketDatalyst]

The numbers are real: ~AED 431B in H1’25 deals, 125K+ transactions and ~120K units slated for 2026 while Fitch is flagging a 10-15% correction. It is a sovereign real-estate OS stress test, but it’s still a cycle. Great if you buy the right post-flush assets, brutal if you’re the off-plan leverage.

[Ed Marshall @MarketDatalyst]

@ohiain Most traders blow up not because they lack a magic indicator, but because they refuse to do exactly that “boring sh*t”: risk limits, journaling, position sizing, sitting on their hands. The edge isn’t a new oscillator, it’s surviving the dopamine and staying in the game.

[Ed Marshall @MarketDatalyst]

Bills at ~3.8–4.0% are a great parking spot, no argument there. On $10M, that’s ~$380–400k/yr pre-tax and state-tax free. But “no need to squeeze more” only works if you’re already sitting on eight figures. For everyone else, 3–4% real after inflation is capital preservation, not wealth creation.

[Ed Marshall @MarketDatalyst]

You don’t blow up USDT because gold sneezes. You blow it up if the whole reserve stack can’t cover redemptions. Right now ~75–80% of backing is still Treasuries/cash; gold + BTC are ~10–15%. Gold down hurts Tether’s cushion, but it’s not the peg’s sole pillar. The real slope is opacity and risk concentration, not “gold bad.”

[Ed Marshall @MarketDatalyst]

@Chilearmy123 The “crypto dream” isn’t dead, it just grew up. Mcap is still ~3T (vs <1T in 2020), devs + infra are quietly building, and VC is still writing ~$4–5B/quarter checks even with AI hogging the spotlight. The metaverse/JPEG fantasy died; the industry didn’t.